SEMClubHouse - Key Relevance Blog » SEO

Exploring Dates On Pages As A Ranking Factor

silver — Wed, 01 Sep 2010 12:10:44 +0000

by Chris Silver Smith

During the past year, I became a little excited at one of Google’s many enhancements to the presentation of search results, because I suspected it could hint at a possible ranking factor they might’ve introduced. The element in question is a date stamp.

You may’ve noticed that in some cases Google will prepend the usual listing snippet text with a date. That change was introduced sometime around late 2008 or early 2009. I noticed the addition of the date with interest, but I became even more interested after I heard Matt Cutts state in a Webmaster Help video that Google considered the date to be helpful to users:

When Google states outright that they consider some element of webpages to be “useful” to searchers, my ears prick up, because Google is so obsessed with Usability that they sometimes use quantifiable elements of user-centered design in their search algorithms, such as their recent introduction of Page Speed as a ranking factor. In this way, Google’s Usability fixation can reveal ranking factors.

I wasn’t alone in twigging to the dates in search snippets — Michael Gray thought their introduction might be significant, Barry Schwartz reported on them, and Performics also shared my suspicion that dates might influence rankings, although their analysis didn’t find any evidence that it did. Edward Lewis (“pageoneresults”) also observed the date snippets and theorized that they might be beneficial for newer content.

On one hand, there’s a compelling reason to think Google might give ranking benefit to pages which have the date — their user testing has indicated that people may prefer to be able to see the date in the search snippets. Therefor, pages which have this data which could be shown in a snippet might be given a small boost in rankings in order to give searchers more of what they want.

Edward Lewis’s early theory also makes sense — for some types of content such as news stories and blog articles about current events (and perhaps even for photos), the freshness could make a difference. Yet, other types of articles are about old subjects or could be timeless — in which case the freshness wouldn’t/shouldn’t affect rankings.

I tested just now while writing this by searching for a major current event from earlier in internet history, and I see pages with 1997/1998 dates in the snippets. See: “worldcom and mci merger“. Since there are even more current articles which mention those keywords, the fact that some of the older stuff ranks higher seems to indicate to me that the actual age is not necessarily a ranking factor so much as perhaps the presence of the date itself.

Another theory could be that while Google might not use the presence of a date as a ranking factor, it could still indirectly affect rank because users might be more inclined to click upon search results where they can see the date value. If CTR from search result listings is a ranking factor that Google uses, then the presence of the date in the snippet could be impactful. (Clickthrough Rate or “CTR” as a ranking signal is a controversial subject, along with bailout rates, which I fully recognize.)

Since I think that there’s a high chance that the mere presence of a date may benefit search rankings either directly or indirectly, I believe it’s better to have them than not, particularly for articles and blog posts. But, this is a point where Michael Gray and I diverge.

He recommends revoking the date after 3-6 months, on the premise that users will tend to be biased against older content in nearly every case. In some cases this is a very valid point. For instance, if I’m trying to research details about some software or hardware, I’ll want current information, and not something six years old. So, for that type of content — Michael’s right — consumers would avoid clicking on the result that appears too stale to answer what they need. The downside of this is that I’m going to bail out of his page ultra-rapidly, if I figure out that it’s old content that can’t help me. If I can’t find any contextual clues to the freshness, then I might waste more of my time there, but ultimately become highly annoyed with the site for failing to post a date, since info about how to do something in software from 8 years back is not going to help me. From this standpoint, suppressing the date field is annoying and just not nice.

But, for date-neutral information, he has a point — adding the date to the SERP snippet for articles which are not date-dependent could introduce a bias against clicking which might not otherwise be there. Assuming that the consumer doesn’t recognize that the date is not a high priority in the context of the particular subject matter.

So, how do you decide what to do about all these variables? Google really hasn’t provided enough information from their usability studies to allow us all to make a decision. In the wake of that, I’ve outlined some possible good reasons to encourage dates in snippets, and also some bad reasons. But, where Michael Gray has decided to eradicate dates in most cases after a few months, I thing there’s enough information to argue that we should include dates in more cases than not.

To illustrate this, I’ve come up with a decision matrix to illustrate why I think there’s more pros than cons in choosing to display dates on one’s site. I created a column for each kind of content — fresh or old and date-dependent or date-independent, and a row for each theory on possible impact of date in SERP on rankings and CTR. The green check marks indicate that the combination would be beneficial, and red ex-outs indicate that a combo would have negative impact:

As you can see, there are relatively fewer possible cases where inclusion of the date would have a negative impact. All of those cases are for older content, but, as I pointed out with the very first screengrab, I see lots of cases where Google provides very good/high ranking preference for crusty old pages which have old dates (perhaps partly due to ranking factors giving preference to older, trustworthy, authoritative content).

Just anecdotally, unless it’s about current events or software details, I know that I click on old-dated content as well as new. So, I’d argue that age doesn’t at all count against rankings, nor even CTR in most cases.

Could Google be using the mere presence of the date as a ranking factor? Maybe. If so, I think that it’s a fairly weak factor compared to many others. But some of these weaker factors are the very hardest to obtain clear research data about — even if you’re an SEOmoz, you have difficulty in isolating the effects of one weak factor versus well over one-hundred other variables.

Due to the matrix I’ve shown above, I think there are more reasons than not to include the dates on your pages. However, I think that Michael Gray does make a number of valid points. You know your own content better than an algorithm, and if you know that your content doesn’t lend itself to display of dates, then avoid doing so. Tread carefully, though, because elements that Google deems important for user-experience may be closely associated with your rankings and CTR. You might ought to actually test whether pages perform better with or without dates before you summarily opt out. Try A/B testing it first.

I’d side with Michael’s implicit stance that Google probably ought to enable webmasters to opt out of display of dates if they choose. I’m generally a fan of giving webmasters rather more control over their own listing display than not — such as what was done with Yahoo’s SearchMonkey and Google’s Rich Snippets.

I’d also strongly encourage Google to publish more of their own internal findings about elements of usability or user-experience. The opacity in this case has resulted in a number of large content sites choosing to suppress dates on pages — a trend that I dislike as an internet user.

Off-Label Use For Google’s Image Labeler?

silver — Thu, 19 Aug 2010 16:26:56 +0000

by Chris Silver Smith

One of the creative methods Google has used for associating keywords with images is their Image Labeler game, which has been in “beta” for some years. As you may be aware, it takes images from their extensive repository of spidered pictures, and assigns one simultaneously to two different people who opt to play the game. Each participant submits keywords describing the image presented to them, attempting to also match keywords submitted by the other participant.

If you’ve reviewed very many websites and webpages, you’ll quickly see that there would be a great many cases where Google might spider some images, yet not have very much data to go on in terms of what the image is all about. Ideally, webmasters add images onto webpages with very clear captions right below them, and also use the ALT parameter in the IMG tag to tell what the image depicts. (ALT Text or “Alternative Text” is a parameter that allows a designer to supply some meta-data with an image — the ALT text describes the image in text, enabling audio browsers to speak the image’s text for blind and vision-impaired web users, and the text can also be used by search engines.) Well-optimized sites might even have their image filenames also reflect descriptive keywords, too. However, it’s frequently the case that a webpage designer neglects to do such things, leaving search engines to try to decipher how to make the images appear for appropriate keywords.

So, Google’s Image Labeler game is one of many methods they’re using to overcome the lack of info they encounter in crawling the web. (They also employ some more sophisticated techniques in combination with this, such as supervised multiclass labeling and optical character recognition (“OCR”).)

It recently struck me that Google could easily make use of the Image Labeler in another way as well — a sort of hidden, “off-label use” of the technology.

Since ALT text is a way of hiding text on webpages, it’s also been misused quite a bit by less-ethical SEO marketers and by people who wisht to take shortcuts in promoting webpages. Search engineers have long warned marketers not to put too much text into the ALT element, or else it could be deemed to be spammy behavior.

So, one way that Google could use Image Labeler would be to sometimes check the honesty of the website designer. Images with suspicious amounts of ALT text could be analyzed via Image Labeler, and if people didn’t submit words that closely related to the ALT text, the image and the site where it’s hosted could recieve a low quality score rating — or even become penalized.

One of the types of images that is frequently prone to keyword-stuffing is a site’s logo graphic. Websites which have suspiciously high amounts of text within an image that has “logo” in the filename could be submitted via Image Labeler, and if the game’s players didn’t validate the keywords associated, it could get the website dinged. Here’s one example of a Dallas plumber’s over-stuffed logo, and its ALT text:

Of course, Google has other ways of detecting keyword-stuffing in IMG ALT text. I’ve heard their image search engineers suggest that ALT text that’s longer than a certain number of words could be considered automatically suspect. Likewise, I’ve seen instances where they appear to be using OCR to automatically convert text embedded in graphics into searchable keyword text.

But, I’m also sometimes seeing some images which have text in them presented in the Image Labeler game, so I think it’s quite possible that they could indeed be employing the “off-label use” I’m describing here.

Facebook SEO Tip: Add Your URL To Your Wall

silver — Tue, 17 Aug 2010 16:41:28 +0000

by Chris Silver Smith

Here at KeyRelevance we’re researching a number of different avenues for online marketing for our clients, so, along with our bread-and-butter work on Paid Search (PPC) management, and Search Engine Optimization (SEO), we’ve done quite a bit of exploration of ideas on how to leverage the massive audiences found in various Social Media such as in Facebook and Twitter.

Yesterday, I published an article on a somewhat subtle technique which can be used when posting status updates on Facebook in order to increase the numbers of people who might see each updates. However, there are a number of very straight-forward things which businesses and organizations can do to extract marketing advantage from Facebook without getting all tricky. Sometimes the most basic steps can give you the greatest advantage, but it’s not always obvious how to go about it.

So, here’s a ridiculously basic tip which I’ve found many businesses have utterly failed to accomplish in setting up their Facebook presence: add your website link to your Facebook wall page!

There are a great many companies, organizations, and small businesses which haven’t figured out how to do this, and so you can encounter pages all the time which do not sport that most basic element of their online marketing. For instance, the official Facebook page for the University of Texas at Austin, one of the largest universities in the country, has completely missed the boat by leaving their URL off their Facebook page:

By contrast, their rivals at Texas A&M University have implemented their website URL on their Facebook page:



(Disclosure: Texas A&M was my alma mater, so I did get a grin when I noticed that the TAMU University Relations Department did this most basic element correctly while the "Tea-Sips", as we like to call them, did not.)
Oh, to be certain, I should point out that URLs on Facebook pages are nofollowed (not to mention that they're apparently dynamically written to the pages onload, via Javascript), so they're not precisely as optimal as many search engine marketing experts might like. However, there's much to indicate that Google, if not the other search engines, can pick and choose whether they'll use the links from a particular site as ranking signals, even if the links there are Nofollowed. Wikipedia is a strong case-in-point, where it appears that links from the behemoth site are somewhat influential. Google might well desire to use Facebook page links as well as the number of likes associated with a page as yet another ranking signal -- despite the presence of the nofollow parameter.
Part of the reason so many companies (or universities) may have neglected to add their links to pages is Facebook's fault. Depending on the type of page one is adding, there may or may not be a field provided to add a link -- which is downright odd. Most people have set up personal pages in Facebook, which allow one a specific field to enter the URLs into -- so, people are at a loss when configuring the page for their business and no link field is provided overtly.
So, one way around this limitation is to update the small text box located below the profile picture that says "Write something about [organization name]." If you write out your URL here, it will automatically get hyperlinked by Facebook. Here's the field from our KeyRelevance Facebook page:



Using either the full, HTTP://... URL, or just "WWW.EXAMPLE.COM" version of a URL will result in it getting automatically hyperlinked once you save.
Ideally, I think one should enter the business/organization URL, and then post a brief descriptive statement directly afterward, similar to how Texas A&M University did theirs.
Anyway, enjoy this most basic of tips -- the link is the most fundamental element of search engine optimization, so you should never neglect including it with your online promotional work. Unless you're the University of Texas -- in which case I'm okay with it.


Using Bing’s New Webmaster Tools For SEO
silver — Fri, 06 Aug 2010 12:05:41 +0000
by Chris Silver Smith
You may be aware that Bing recently released a new version of their Webmaster Tools which are intended to help webmasters in improving their sites’ performance in Bing search. One of Microsoft’s Senior Program Managers and SEOs, Duane Forrester, asked a number of us to give feedback to their team on what could be improved about the interface. So, I thought it might be good to provide that feedback via blog post, openly — not to beat up on Bing, but to further bounce ideas among the community.



Giving feedback to any of the search engines about their tools for webmasters seems a bit fraught with the near-futile dichotomy between the desires of Search Engine Optimization experts and the desire of search engineers to neutrally provide positive/fair rankings of search results. However, the exercise of me providing a little feedback is worthwhile, because if the tools are useless or pointless to us, then there’s little point in the search engines going to the effort of providing them in the first place.
Having worked in a major corporation before, I almost feel repressed about throwing out suggestions that I know could be deemed no-gos from the point of view of Bing engineers. I tend to self-censor to a degree because I don’t want to be interpreted as naive of the issues the search engines must take into account in trying to limit undue influence of those attempting to subvert the SERPs.
Even so, I’m aware of the potentially conflicting considerations, and as I described earlier, it’s an exercise in futility if the tools don’t provide worthwhile functionality to the intended users.
One of the primary problems I see with Bing’s Webmaster Tools is the sense of “keeping up with the Joneses” one gets when reviewing their interfaces. Bing’s development team is in a near no-win situation with whatever they do in this area. On one hand, if they copy the same functionality found in Google’s Webmaster Tools, they’d be accused of being mere immitators. However, there are some good elements in Google’s toolset which really ought to be provided, perhaps. On the other hand, if they went even further in providing usefulness to webmasters, it could make them more prone to unethical marketing exploits. So, there likely were not a lot of easy solutions nor perhaps obvious things which they should have done.
Further, their focusing upon their tool-vs-Google’s tends to be a bit incestuous, and there’s the usual engineer myopia in providing what they think people would need/want versus trying to really look at the problem directly from the point of view of a webmaster. (Now, this bias in perception can’t be accused of Duane, because he was an external SEO prior to working for Microsoft — but there’s a definite sense of this basic utility design problem inherent in both Bing Webmaster Tools as well as Google Webmaster Tools.)
Likewise, Google Webmaster tools suffers a bit from the conflicting goals of the engineers and the needs of the tools’ target audience. So, I’d prefer that none of the search engines look at one another’s offerings when designing such things, but instead try to focus solely upon providing as much functionality as webmasters might need. As things currently stand, there’s a sensation that all of the search engines are providing something of “placebo utilities” to webmasters — the interfaces have some confusing melange of features which are ultimately not all that useful, but are instead intended to throw up some smoke and mirrors to make it appear that they’re trying to help webmasters with the optimization of their sites.

Moving past my perhaps-unfair assertions, let’s look at what the new Bing tools provide, and what could be done better.

First, a head-nod to Vanessa Fox for her comparison between Bing's and Google's Webmaster Tools -- as the creator of Google's Webmaster Tools, she is likely one of the best people around to examine such utilities with a critical eye, and in the best position to know how much info a search engine might realistically be able to provide, and in what format. Likewise, a nod to Barry Schwartz's post about Bing's tools.

Both Vanessa and Barry berate Microsoft for building the tools while requiring Silverlight technology to view/use them. I don't consider that as much of a big deal, because I consider that a sort of "religious difference" in how the tools were constructed (most of us who are jaded about how Microsoft has strong-armed proprietary technology in the past might react negatively against Silverlight, as well as those who avoid it out of conservative privacy practices).

However, if I'm looking at Bing Webmaster Tools purely from the perspective of how well it does or doesn't function, I'm not concerned about this tech dependency built into it, since I think the majority of webmasters out there will be unaffected by this. I'm not a fan of MS programming protocols AT ALL, and it may be a bit of my former bias as a technologist within a megacorporation creeping in, but the Silverlight criticism just appears slightly out of sync with the primary issues of whether the tools provide vital functionality or not -- and, it may not be unfair of Microsoft to decide that if you wish to play in their Bing sandbox, they have the right to promote their proprietary technology to do so. In comparison, I have a friend who is a privacy freak, and he surfs with Flash disabled -- Google's Webmaster Tools requires Flash for one or two graphs, and would be equally irritating to him as Silverlight.

Both Barry and Vanessa mention how Bing's new interface revoked the backlink reports, and I agree with them both on this point. This was one area where I'd hoped Bing would take the opportunity to be more open than Google. If the engineers looked at competitors' tools while building Bing's, they should have tried to recreate the backlink reports that Yahoo! provided in Site Explorer -- which seems to give a more comprehensive picture of backlinks. Since webmasters are told that inbound links are one major criterion for rankings, avoiding providing this info is a major void.

Bing obscures the numbers of pages indexed when one performs a "site:" search by domain, too, so revoking this functionality, such as it was, from the old interface eroded some of the usefulness. Perhaps their pre-development surveying of webmasters resulted in feedback that their earlier backlink report "wasn't useful", but that would've mainly been because it was less-robust than one like Yahoo's.

Vanessa mentions that they don't provide data export features, and I agree completely that this is a major oversight. In fact, as a programmer I happen to know just how relatively easy it is to code data to export in XML or CSV, and considering how long it took to launch the product it's sort of shocking they didn't include this upon launch. (You'd think Microsoft would not miss an opportunity to provide a "click to export to Excel" button!)

Vanessa stated that they also ditched the Domain Score, and remarked that this was a good thing. I disagree on this point because I think any insight into a ranking score that any of the engines give us is helpful in assessing how effective/important a site or domain is. Was this the same as the small bar-scales Microsoft had been providing for a handful of the more important site pages via the interface? Although these graphical page ranking scores were entirely derivative of Google Toolbar PageRank, I would've prefered they provide even more in that area. Bing's in a position where they ought to be able to experiment with providing more info than Google does, and see just how dangerous it really is to be more open with marketers!

Vanessa did a great comparison between the analytics Bing provides versus Google Webmaster Tools and Google Analytics. While analytics from Bing's perspective are interesting to us all, she notes one aspect that also strikes me as an issue with the graphs: as a webmaster/SEO, when I see indexation decreasing, I'd really like to know why. This is particularly irritating where Bing is concerned, because among the industry I think it's widely felt that Bing simply indexes a lot less than Google.

Many of my clients want to know what they can do in order to increase their indexation with Bing. I see the same thing with my test websites. I may have 30,000 discrete pages and Bing appears to index a sharply lower number than Yahoo or Google. The feature allowing one to manually submit URLs seems to acknowledge this sad fact -- but, in context, it's nearly sending the wrong message! "Oh, our spider's legs get tired out there, so bring your pages directly to us." Vanessa's got a point on this score -- why should I feel I need to do this if you accept Sitemaps? And, if I or my clients have tens of thousands of site pages, fifty pages to be manually submitted at a time is simply not a sustainable solution. I can understand having the interface to rapidly submit brand new content pages, but what's missing may be some clear communication as to what issue is restricting my indexation.

The Features showing whether there are robots restrictions, malware, or crawl errors which could impact a site are all great. However, if one already has everything functioning just fine, the tools need to answer futher questions: Why isn't my site crawled more deeply? And: Why don't my pages rank higher? Ultimately, webmasters ask: What can I do to improve my site's performance? Understandably, Bing and other search engines are reticent to provide too much info in this area. However, there are things which they could provide:

Possibly a tool where a webmaster could select one of their pages or submit a URL to find out what primary keyword phrase Bing considers the page to be about?
Tools which report upon quality issues detected with specific pages. For instance, is the page missing a Meta Description or Title? Are there pages where the Title or Meta Description appears to not be relevant to the page's content, or out of sync with anchor text used in inbound links? Are there images which could be improved with ALT text?
Why not merely inform webmasters that you consider their links and other references to be too few or of too low in importance?
Bring back the scales showing page scores, and actually go further in providing some sort of numeric values!
Actually show us that you spider all pages, even if you opt not to keep all in your active index! This would at least give the impression that you are able to index as deeply as Google, but choose not to display everything for other reasons.
How about informing us of pages or types of pages (based upon querystring parameters, perhaps) which appear to have large degrees of duplication going on?
Tie-in our Webmaster Tools with local listing account management for local businesses, so everything could be done via one interface.
Provide means for us to customize the appearance of our SERP listings a little bit, similar to Yahoo SearchMonkey and Google's Rich Snippets.
Provide us with tools that help us improve overall site quality, such as if you see pages on our site with incorrect copyright, misspellings, or orphaned pages.
Consider providing us with an A/B testing type of tool so that you might inform us about which of two page layouts performs better for Bing search!
Inform us if you detect that a site has some sort of inferior link hierarchy -- this could indicate usability problems affecting humans as well as spiders.
Provide more granular details on how well we perform in Bing Image Search, Mobile Search, Video Search, etc. Currently, I cannot tell how many of my images are indexed in Bing.
For that matter, it would be nice to enable an Image Sitemap file, like what Google offers.
Finally, for a really pie-in-the-sky request: You operate web search for Facebook -- would your contract allow you to tell us how often our webpages appear in Facebook search results and how many clickthroughs we might get from those?

Anyway, there's my feedback, criticism, and some ideas for additional features. I'm not trying to beat anyone up and I'm actually grateful for any feedback we receive from all of the search engines on our performance within them. I mainly ask that the search engineers will keep in mind that we mainly want to know "What can we do to improve performance?" and to provide us with tools to accomplish that in as much as they're able to do without compromising the integrity of the engine.

Where Bing is concerned, I believe it could be possible to be even more open than Google is in order to further differentiate yourselves and experiment to see what's really possible in terms of openness!

Keyword Research for Local SEO

silver — Mon, 02 Aug 2010 20:34:36 +0000

by Chris Silver Smith

Doing keyword research for Local SEO has been somewhat difficult in the past, because many local search phrase combinations have relatively low volume, and the amount of data has been too small/granular for the limitations found in many keyword research services.

Even just a few years ago, I used to try to research local keyword phrases for things like “boston plumbers” in a service such as comScore’s qSearch tool, and such phrases would frequently have insufficient search volumes for the tool to reflect back any data. Even Google Trends today states that there’s insufficient volume to show graphs for “boston plumbers“.

The reason why such research is important is because a site seeking to grab as much qualified traffic from consumers interested in a particular type of business as possible, must first know which phrase(s) to focus upon to achieve search engine rankings. They must answer the questions of whether consumers are searching for “plumbers”, “plumbing”, “plumber” or “pipes”. And for local businesses, it’s ideal to match for exact phrases that include local keywords. Do consumers search for “boston plumbers”, “plumbers, boston”, “plumbers in boston”, “plumbers boston ma”, or “plumbers 02118″ (a Boston ZIP code)?

There are cases that are even more complex, where an industry may have multiple terms used to find businesses (“accountants”, “accounting”, “tax preparation”, “CPAs”), and cities with multiple name versions and neighborhoods (“New York”, “New York City”, “Manhattan”, “New York, NY”, “NYC”, etc).

The problem is irritating when there’s little recourse available for researching consumer behavior.

However, various providers have been beefing-up the data they make available in order to help address the marketing demand. For instance, Google Insights will sometimes show us the relative search traffic for phrases in cases where Google Trends will not, such as for the “boston plumbers” example:


(Click to view larger)

One reason why Trellian’s Keyword Discovery tool has been so widely-regarded is that it has long provided very granular information for these sorts of searches. However, their overall data stores are less-comprehensive than Google’s usage logs, so there’s greater likelihood of getting some skewed figures for phrases with smaller volumes. Here’s the “Boston Plumbers” example in Keyword Discovery’s main interface, using the historical database:

As you can see, Keyword Discovery is showing that the phrase with the highest volume of searches that contains both “plumbers” and “boston” is “average annual wage for plumbers in boston mass” — which is somewhat surprising. One figures that there are likely a lot more consumers trying to locate plumbers than there would be plumbers looking up wage info or prospective employees of plumbers. There are multiple possible explanations for this surprising result. For one, it could be that someone has linked-to or set up a script to frequently execute searches for this info. For another, there could just be an unusually high number of those searches happening just in the sample of data that Trellian has access-to.

Trellian and other keyword research providers obtain usage data from ISPs and some search engines (typically some of the lesser search engines) which they compile. Since this is a subset of all internet usage data, it’s good to realize that searches with smaller volumes like our example can skew results a bit — so, one must interpret it with a critical eye.

The results also show a number of search permutations involving “local 12″ and “local 112″. These are unions, and those searches probably were not from individuals seeking plumbing services.

Two of the remaining search phrases we see are for “plumbers list boston” and “boston area plumbers”. Both of these are interesting/worthwhile results, particularly since we didn’t initially anticipate those phrases. That second phrase, “boston area plumbers”, is likely of prime benefit to us if we were a local plumbing company. Boston, like many major metro areas, is made up of a conglomerate of smaller towns, so it might be very natural for locals to search “boston area plumbers”, knowing that while there might not be a plumber located directly in their specific town, there could be one in the town adjoining theirs.

5minutesite.com’s Local PPC Adwords & Keyword List Creator by Walter Spurgiasz is pretty good for generating a large list of permutations of local keyword phrases. You submit keywords, a ZIP code and radius, and it will incorporate names of towns within the searched radius and add them onto your keyword phrases. Another similar tool is the Local Keyword Research Tool. These tools are good for coming up with phrases and ideas, but they’re not going to tell you which word/phrase is more popular than another.

In the case of our “Boston plumbers” example, though, these tools could be very useful since they’d spit out a great many of the local town names which you’d want to also target for local searchers.

Ultimately, we may find that we might need to target phrases which include town or ZIP code names less in the future, since search engines such as Google are increasingly returning back local results for broader keyword phrases, targeted to consumers’ geographies (by using IP addresses and a number of other geotargeting techniques). For instance, when you search for just “plumbers”, Google now returns back a handful of organic search results PLUS a map 7-pack of local businesses in your area which are plumbers. This evolution in search engine function is likely to alter searchers’ behavior as they grow accustomed to it — they’ll waste less time in typing in their city names for a local business, so the volumes of searches with the longer local phrases will decrease.

If and when this paradigm shift occurs, one of the best tools for keyword research will be Google Insights, once again. Google Insights allows one to submit a keyword such as “plumbers”, and then drill down to see the relative search volume for it by country, states, and major cities:

Trellian Keyword Discovery also allows one to filter down geographically, but the refinement is insufficient — only allowing one to refine by country.

This feature is particularly important for local search optimization, because average searcher behavior across the country changes by regions. For instance, most consumers across the country search for “Chinese Food” when seeking places to eat Chinese. However, in Hawaii, the searchers there search for “Chinese Restaurants” slightly more often:

Good keyword research requires some experience and intelligence, coupled with the ability to imagine how consumers attempt to find your business online. I’ve often found that companies have become somewhat biased by terms they consider to be important to their own industry insiders versus the keyword terms that consumers actually use to find them. Imagine the case of a prestigious tailor who desires to rank well for “bespoke tailoring” whereas he’s ignoring the many consumers who might find him if he targeted “custom suits” instead. Periodically revisit assumptions about how consumers may search for your type of business, and avoid vanity terms.

Matt McGee also wrote some great tips last year on this subject at When Local Keyword Research is a Dead End – it’s worth reading as well.

Yahoo! Now Mirrors Bing Results – SEO Reflections

silver — Thu, 29 Jul 2010 21:21:28 +0000

by Chris Silver Smith

As you may be aware, Bing is now powering the Yahoo! search engine.

We’ve been watching this development since it emerged and comparing results.

For the most part, the organic keyword listings on Yahoo SERPs match up with Bing’s one-to-one, but we do see a few notable differences.

Thus far, we’re not seeing Microsoft PPC ads on Yahoo, and the local map results are different. Yahoo appears to have opted not to display Bing’s groups of related search results or the city-specific block block which can be found at the bottom of most Bing city SERPs. This is most obvious in the results for things like “Dallas, Texas.” This is good in the sense that Yahoo’s first page remains a set of ten results related to your keyword (as opposed to Bing, where page one may show only a Top 5).

Meanwhile, Bing/Yahoo’s combined search share continues to rise, and many have been seeing ongoing increases in Bing referral traffic to their sites.

With Bing search share rising, and their SERPs now being mirrored on Yahoo, the question prime on many marketers’ minds will be: What should I be doing for Bing SEO and Bing Local SEO?!?

For some years now, the Microsoft/Bing (and even Yahoo!) SEO advice has pretty much been: “do what you do for Google, and it works pretty well for Yahoo and Bing”. This advice has held up pretty well, particularly on the general assumption that since Google’s marketshare is so large — we pretty well must aim for the greater target and pick up referrals from the lesser sources as a nice bonus where we can get them. However, as Yahoo and Bing search traffic combines, more focus is going to be placed upon differentiating factors between the Google/Bing search giants.

In our agency, we have increased focus on where there may be some factors which are ranked more heavily in one search engine versus the other. Just a few observations for publication thus far (particularly focusing upon organic rankings for local queries):

Bing may show preference for shorter URLs.
Better-known name brands appear to have more advantage in Google than in Bing. (This difference suggests that Google may be incorporating more non-link citations, such as mere mentions of bizname in text, into the ranking mix, while Bing may not be using such a signal. Although, many have noted an overall bias on Google’s part for name-brands above lesser-knowns.)
There’s a greater degree of duplication in Bing SERPs for local queries, in the sense that the same domain may be allowed to appear more than once in the same keyword SERP.
Unsurprisingly, Google appears to be doing more to normalize URLs, trying to limit dupes caused by things like uppercase/lowercase variants of the same page from diluting out in the index.
Bing appears to prefer biz profile pages more, compared with Google which prefers higher-level, larger-PageRank directory pages. But, Bing prefers directory sites more than Google overall. (This could change even more dramatically in this direction if Google’s Local Search “Experiment” goes out of testing into wide release.)
Bing may prefer to rank a website’s homepage more than deeper content.

Rand’s “Google vs. Bing: Correlation Analysis of Ranking Elements” at SEOmoz and Aaron’s “Optimizing for Bing” post appear to have fairly solid observations which align with some of what we’re seeing as well.

We’ll be further focusing upon Bing local ranking elements, and I might publish further observations in the near future.

Google Penalty For Low-Quality Writing?

silver — Mon, 21 Jun 2010 12:01:49 +0000

by Chris Silver Smith

For a while now, I’ve been covering how Google’s increasing focus upon quality measurements are steadily translating into actual ranking factors. Four years ago, I first conjectured that Usability could supplant SEO. Back then, we could see that Google’s human evaluators added quality ratings into the mix, affecting page rankings. Since then, Google added helpful tools for usability testing and page speed diagnostics. This year they’ve continued this progression by incorporating page speed as a ranking factor and the recent “Mayday Update” apparently shifted some ranking factor weighting from keyword relevancy to quality criteria.

Considering Google’s desire to quantify and assess elements of quality in webpages, what are some other possible things which they might attempt to algorithmically measure and base rankings upon?

One possible area which occurs to me is in testing the text body of pages, particularly that of the main body of articles and blog posts.

Spammers frequently use programs to automatically assemble snippets of text for the purpose of targeting many combinations of keyword phrases without requiring that their pages be all hand-written. Some spammers merely steal others’ content, screen-scraping pages and redisplaying it on their own sites. The more savvy ones know that the search engines seek to detect duplicate content, and try to credit the originating sites as authoritative for matching keyword phrases. Recognizing that purely identical text can get filtered due to duplication detection, these spammers may automatically insert random words throughout the text, resulting in weird sentences and nonsensical writing.

Less dramatically, marketers who desire to rapidly develop out thousands of pages of content sometimes resort to copy writing companies that outsource article assignments to third-world countries. Poorly-educated writers result in terrible grammar and bad spelling. And, foreign companies sometimes hire bad translators to convert their pages for English readers. (Such bad writing can be entertaining — check out Engrish for samples to make you grin.)

Spammy sites and pages with poorly-written articles would definitely be deemed to be low-quality by most consumers. Most of us don’t want to end up on such sites, and Google doesn’t want to lead us there. So, if I were them, I’d try to find ways to detect such poor content.

But, would it be possible for Google to detect bad writing?

I think the answer to that is a resounding “yes”!

Have you noticed how many different software packages offer spellcheck functionality? And, software such as Microsoft Word can assess grammar in documents as well as word/phrase variety and reading level.

So, it would be possible for Google to detect bad writing. I’m not sure how process-intensive this sort of functionality would be for them, but it could easily be incorporated as a subsidiary process that operates after many of the more-rapid ranking systems have already assessed a webpage. A grammar/spelling grading process could trickle-in ratings over time.

Is Google using such a process?

For now, this is unclear. It would be hard to isolate the effects of such a process, since their human quality evaluators might also negatively rate a page for some of the same reasons. A few different search engine patents relating to display of search snippets, assessing reviews, and ranking of news stories mention the possibility of utilyzing grammar and spelling in ranking processes. This sort of text analysis appears so simple that I think it would be a no-brainer to believe that Google would use such a system.

For spammers, I’d say that the bar is moving higher. Ignoring quality for the sake of a fast buck is getting more difficult. The days of merely sprinkling a keyword phrase randomly among paragraphs of words are probably coming to a close.

For marketers, the implications of text quality assessment should be clear. Cheaply-written content ought to be avoided for sustainable, long-term benefit. And, if you wish to improve your rankings, have a professional writer with good English skills look over all of your content, and correct any errors they may find.

You don’t really even need proof that this could be used as a quality ranking factor for it to make sense to clean up the grammar and spelling on your site. Poor grammar and spelling can make a bad impression on consumers, resulting in loss of trust and lower conversion rates. Improving your site’s user experience provides short and longterm benefits.

The Mayday Update Skinny From SMX Advanced

silver — Mon, 14 Jun 2010 12:03:56 +0000

by Chris Silver Smith

At the SMX Advanced conference in Seattle last week, the keynote session with Matt Cutts has become an expected feature, but it’s also one of the most highly anticipated and attended sessions of the entire conference. The search engines love to take advantage of search marketing conferences to make major announcements, and Matt Cutts has been known to drop both major and minor bombshells during these sessions. For instance, during last year’s session, he stated that the practice of “link-sculpting” (using “nofollow” parameters on links to advantageously design the flow of PageRank within a site) was now pointless, because Google had implemented nofollow such that it did not conserve PageRank, but instead a nofollowed link merely evaporated PR.

It was clear at this year’s “You & A With Matt Cutts” that he and Danny Sullivan had planned in advance to launch directly into addressing one of the top most-recent issues of interest to webmasters: the “Mayday Update” — so-named because the algorithmic shift occurred roughly around the first week of May, and because affected webmasters were left with a helpless feeling after their pages dropped in rank for long-tail search queries.

Matt and Danny opened the session in a really jocular fashion by wearing inflatable life-jackets, as a nod to the Mayday algo change. They followed that up by handing each other caffeine-free sodas, which they quickly deprecated in favor of fully-caffeinated Coca-Colas. (As you may know, Google began rolling out an infrastructure/processing change this year, called “Caffeine”, which allows Google to rapidly absorb fresh content, process it for ranking purposes, and display the new content in SERPs. Some webmasters who were concerned over the Mayday Update had wondered whether it might have been caused as some side effect of the Caffeine change.)

After the lifejackets and soda hijinks were over, Matt stated clearly and seriously that the Mayday Update was separate from and in no way caused by the Caffeine change. His statements further underscored statements he’d made earlier online. According to him, the ranking algorithm development team had decided, after consideration and testing, to publish a change based upon some “quality factors”, reducing the rankings of some deeper content pages for longer-tail queries.

Just as Vanessa Fox had opined in her piece on the Mayday Update, the weighting for keyword relevancy factors was likely reduced some in comparison to quality factors.

One thing that Matt suggested to those who wished to counteract Mayday’s effects was to redesign a site’s links so that they wouldn’t be as deep within the site structure. (From chatting with close colleagues at the conference, Chrstine and I also have a few other strategies for counteracting Mayday effects, which we intend to keep in confidence for the benefit of our clients only.)

Another hot-button area that Matt and Danny discussed was that of the supposition that Google AdSense advertisers might get superior page rankings, because of the suspicion that pages linked-to from ads might receive better document analysis from the ads’ bot/processor than from the processing of strictly organic pages. This discussion came up due to a question about how well Google could interpret JavaScript content. Matt tried very hard to answer the question comprehensively, but his effort to be very specific in answering anticipated qualifications to the question gave many in the audience the sensation of a politician trying to side-step a delicate question.

(My opinion is that he answered this question truthfully, to the best of his knowledge, but his overly careful and hesitant phrasing made it come across as someone trying to provide a limitedly-true statement while dancing around the intent of a quesiton — rather similar to how one must carefully phrase a wish to a genii in the bottle or else get the wish answered in a literal-yet-unintended fashion. But, you have to cut the guy a break on this — when he answers a question with a flat-out, simple answer, people dissect it to death, searching for exception cases or conspiracy. So, no wonder he’s trying to carefully and specifically anticipate every permutation of the question and answer!)

Throughout the conference, Matt tried to further emphasize how marketers should try to pursue user experience, since that’s what the search engines themselves are pursuing, instead of obsession upon ranking factors. He suggested that ranking factors change over time and have a more limited lifespan than usability and user-experience criteria, so user-centered design is far more sustainable over the longterm. The continued focus on user experience is gratifying to hear, as it’s a philosophy we’ve endorsed for a long time as well.

Title Tag Tips To Get To The Top

silver — Thu, 20 May 2010 15:48:51 +0000

by Chris Silver Smith

One of the most basic, and effective ranking signals used by all major search engines is the text put into a page’s TITLE tags. The element is so foundational, and so strong that it’s one of the first things that any organic search marketer recommends for non-optimized sites, yet few people know how to do it really effectively. Here’s a few tips for formulating titles that work for a great many common types of webpages.

As you may know, the HTML Title of a page is typically found at the beginning/top of the page’s formatting code, within the tags. The text within the tags is only visible when viewing a page at the very top of the browser window. <code> <div align="center"><a href="http://www.flickr.com/photos/silvery/4623916507/" title="KeyRelevance's Title Tags by Si1very, on Flickr"></a></div> </code> This text is also frequently used by the major search engines as the link text for the page when it appears in the search results. As the title of the page, it’s supposed to succinctly represent what the page is about. The W3C <a href="http://www.w3.org/QA/Tips/good-titles">declares</a> it to be the most important element of a quality webpage, and recommends that it not be too short nor too long. Common mistakes that webmasters do include making all of a site’s titles identical, and using terminology in title text which is unlikely to accurately reflect the page’s content or which isn’t synchronous with what consumers are using when they type queries into search engines. There is some degree of art and experience that goes into creating really effective Titles, but here are a couple of fairly common formulas that work well, depending upon the site and the webpage: <ul> <li><TITLE>Keyword Phrase + Brandname

1st Keyword Phrase, 2nd Keyword Phrase, Possible 3rd Keyword Phrase + Brandname

The first example is very tightly-focused. This format would be used when one has moderately good confidence that the webpage can rank very well for a specific keyword phrase. By only using one keyword phrase, the page is made to appear much more specifically relevant for the keyword topic.

For instance, I just performed a search in Google for “Nikon Coolpix S3000 camera”, and one of the top-ranking pages for it was from NiconUSA.com website with title text of “COOLPIX S3000 from Nikon”. This title is simple, tightly-targeted to the keyword phrase that consumers are searching upon, and it’s well-formed with the keyword phrase placed first, before the brandname, lending it greater importance/weight in the search engine’s algorithmic assessment.

The second example involves including two or more keyword phrases within the Title, and this would be used when one decides that there’s perhaps a lesser chance of ranking high for the main, most-popular keyword phrase. This formula is broader, including one or more secondary keywords or keyword phrases — casting the net wider to try to collect more long-tail traffic. (This format could also be used if one is highly confident of retaining a good position on the main keyword phrase, but expanding to target additional phrases as well in order to rank better on those secondary terms.)

The decision process for using a broad versus narrow phrase formula in Titles is made by conducting keyword research and by carefully analyzing the competitive landscape for the keyword phrases involved. In some cases, there may be little competition for ranking very high for a phrase, but this situation happens less and less as more webmasters have become aware of good design principles and SEO.

There are quite a number of other page Title formats, depending upon the page type and content involved.

Homepages typically involve presenting the brandname very well while also targeting the top most-important keyword phrase for the company, and a phrase which describes what the company is or does.

Locally-oriented webpages need titles which not only include a keyword phrase and brand, but also terms which specify the locality.

Particularly lengthy keyword phrases or brand names sometimes require special handling and clever arrangement to try to keep the title short enough to be readable within search results while also matching upon consumer search queries.

Finally, some Titles are also dictated by common nomenclature used on websites, such as for “About Us”, “Contact Us”, “Privacy Policy”, and more.

For some limited feedback on your page titles, be sure to register for Google’s Webmaster Tools, because they provide recommendations under “HTML Suggestions” in the “Diagnostics” section, alerting you when they encounter missing or duplicate title tags. They also tell you when their algorithms assess page titles to be too long, too short, or “non-informative”. Use this feedback to further fine-tune your titles.

If you’re a do-it-yourselfer, follow the general principles I’ve outlined and you may see a significant improvement in search engine rankings and traffic.

Without Usability, You’re Not Doing Advanced SEO

silver — Tue, 27 Apr 2010 16:06:14 +0000

By Chris Silver Smith

My article covering how Google’s fixation with Usability reveals local search ranking factors published yesterday on Search Engine Land. In it, I described a number of common website elements which few-to-no marketers have ever cited as ranking signals. Some of these elements, such as whether or not a site may have employee profile pages, or whether a site displays prices for products and services offered, might be controversial in search engine marketing circles.

CNN's homepage checked with Google Page Speed - Google introduced Site Speed as a new ranking factor in 2010, and provided tools like this Page Speed extension in FireFox to assist webmasters with Usability improvements.

Other elements I described have been cited by other experts as beneficial for search marketing, even though they may’ve recommended them for reasons other than those I outlined. Inclusion of images, maps and locations pages make sense for multiple reasons in local business websites.

The thought and methodology behind coming up with these factors is sound, and has allowed me to successfully predict present and future search engine optimization factors where others have not. It makes logical sense that while Google is interested in Usability, they will seek ways to quantify and measure it on websites, just as they have done with Site Speed. And some very easy usability elements to quantify include common website elements such as the About Us, Contact Us, and Locations pages.

Back in 2006, I began predicting that the practice of Search Engine Optimization might become replaced by Usability. Unquestionably, this change is occuring to some degree right now.

I’ve known a lot of top corporations which are involved in very sophisticated paid search marketing and search engine optimization, but few of them are also including usability testing and user-centered design considerations when performing a site redesign. Google has tried to make the importance of user-experience abundantly clear by actually going public with their adoption of page load times in determining search result rankings, but many companies are still not connecting the dots.

Here at KeyRelevance, we have long prioritized usability in our assessments of web sites’ design. When companies contract with us to audit their websites, we offer both a Technical Website Review as well as a Usability Review. However, many companies eschew our Usability Reviews or dismiss them as less-important.

For some reason, people often react to usability recommendations from experts in an emotional way, rather like how a portion of the population avoids going to their doctors for a yearly physical. For some companies, there are already so many dependencies and requirements going into web design projects that they can’t include more without losing impetus. For others, individuals with authority over projects have egos which do not want to lose discretionary control over project decisions which could be altered if usability research ran counter to what they desire to do.

Usability testing can be the difference between a design that becomes highly popular versus one which is rapidly forgotten. Google itself is an example of how user-centered design will translate into success. More design options can be scientifically decided, honing down to interfaces which will maximize ease-of-use and enjoyment-of-use. Instead of being avoided, usability testing should be embraced — after all, in the business world we’re looking to increase the potential for success in our company projects, right?

Knowing Google’s heavy focus upon usability factors, consider that if you’re not doing iterative Usability testing and adjustment for User-Experience, you really may not be doing “Advanced SEO”.

If you’d like a thorough Usability Audit of your site, contact Key Relevance today to schedule our review and get a report of items to consider before your next sitewide redesign is completed.

Also, check out some of the free tools that Google has been providing to help you with portions of usability analysis. Try out Google Browser Size, Google Page Speed, and look at the Site Speed reports in Google Webmaster Tools for your website.